10 replies to this topic

[Lord Method Man]

I've been experimenting with visual C++ having worked in Java for a while. Right now I'm working with a little String manipulation.

I have a Windows Form with a textBox that displays the value of a String. I'm trying to append a char to the String. The problem I'm having is the ASCII value of the String is appended instead of the char itself. So far I haven't been able to find a solution through the resources I've read. Here is an example:

String^ mystring;
char y = 'd';

mystring = "Test";
mystring += y;
textBox->Text = mystring;

The output is "Test100" instead of "Testd" (100 being the ASCII value of d). What should I be doing differently?

[Hardcore Til I Die]

Maybe try using a string for "y" instead of a char?

[dannyres]

I think you will need to do something like this:

mystring += "" + y;

I don't have a compiler in front of me, but you could also try casting y to a string e.g.:

mystring += (string) y;


EDIT: Whoops, just realised this is Visual C++, not C#. Not sure if either of these things will work.

[xorangekiller]

It may be that the reason you're seeing the decimal value instead of the ASCII character is because char is a decimal type. It is a decimal type just large enough to hold an ASCII character: 8-bits. Try referencing it as a pointer instead of a single character to make it a C-style string.

[+Billaguana]

Java programmer myself so walking blindly into a room full of knives here, but is there a method like the one in the String class that you can use to append two strings? Then you just make the char into a string.

[xorangekiller]

Is 'String' a Microsoft-centric data type or did you mistype the name of the standard 'string' class? The string class in the standard library can append single characters, C-style strings, and standard strings using the '+=' operator.

Without knowing the data type of 'textBox->Text', I'm assuming that something like the following will work:

std::string mystring;
char y = 'd';

mystring = "Test";
mystring += y;
textBox->Text = mystring.c_str();

Alternatively, you could forgo the niceness of the string class and use only C-style strings. Your example would then look something like this:

char mystring[50];
char y = 'd';

strcpy( mystring, "Test" );
strncat( mystring, &y, 1 );
textBox->Text = mystring;

[+Asik]

String^ mystring;
char y = 'd';
mystring = "Test";
mystring += y;
textBox->Text = mystring;

This is C++/CLI. The caret "^" on a type name is not C++ syntax, it denotes a managed type, here System::String. If you want to learn C++, make sure you create an empty C++ project, not a "CLR" project. You won't be able to work directly with Winforms with C++; if working with Winforms is what you want, learn C# instead. I strongly doubt you want to learn C++/CLI, it is more complicated than you can imagine and it doesn't serve much purpose besides building bridges between the native and the managed world.

Anyway, with System::String you can't append a char to a string using the += operator, however you could call ToString() on the character and append that isntead, i.e.

mystring += y->ToString();

But really, run from C++/CLI while there is still time. File -> New -> Project -> Visual C++ -> General -> Empty Project. Now you're doing real, ISO C++. It's complicated enough by itself.

[pickypg]

.NET uses unicode characters. Use wchar_t instead of char.

wchar_t y = L'd';

It most likely chooses int (System::Int32) as the closest conversion for char as a result, resulting in the decimal number.

If you really want to use char throughout your code, then you can do something similar to Dr_Asik's suggestion by using Convert::ToString:

char y = 'd';
String ^value = Convert::ToString(y);

This link explains the mapping of types to the CLI: http://www.c-sharpco...arp-programmer/ , which is why it maps to a number rather than a character (it does not widen to the number, rather it widens to an integer).

[Lord Method Man]

Thanks for the tips, Ill try them tomorrow. I'm working on a random password generator. It's something I try to make with each programming language I work with.

[Lord Method Man]

Scrapped everything and redid the project in C#. Holy crap was it easier. I was thinking I could do it in C++ since that's what I used in college (command line progs only). That Visual C++/CLI stuff sucked. Knowing Java I was pretty much able to code in C# without having to look up much of anything. Came up with this little Class for creating a password:

class SecurePassword
    {
	    private String password;
	    private int length;
	    RNGCryptoServiceProvider rng = new RNGCryptoServiceProvider();
	    Random random = new Random();
	    public void generatePassword(bool useUpper, bool useNumber)
	    {
		    char ch;
		    int i;
		    int type;
		    password = "";
		    for (i = 0; i < this.length; i++)
		    {
			    type = (random.Next(0, 4));
			    if (type == 0 && useUpper)
				    ch = getRandomUpperChar();
			    else if (type == 1 && useNumber)
				    ch = getRandomDigit();
			    else
				    ch = getRandomLowerChar();
			   
			    password += ch;
		    }
	    }
	    public char getRandomDigit()
	    {
		    byte[] byteCh = new byte[4];
		    double range;
		    uint intCh;
		    rng.GetBytes(byteCh);
		    intCh = BitConverter.ToUInt32(byteCh, 0);
		    range = intCh / 4294967296.0;
		    intCh = (uint)(range * 10);
		    return Convert.ToChar(intCh + 48);
	    }
	    public char getRandomLowerChar()
	    {
		    byte[] byteCh = new byte[4];
		    double range;
		    uint intCh;
		    rng.GetBytes(byteCh);
		    intCh = BitConverter.ToUInt32(byteCh, 0);
		    range = intCh / 4294967296.0;
		    intCh = (uint)(range * 26);
		    return Convert.ToChar(intCh + 97);
	    }
	    public char getRandomUpperChar()
	    {
		    byte[] byteCh = new byte[4];
		    double range;
		    uint intCh;
		    rng.GetBytes(byteCh);
		    intCh = BitConverter.ToUInt32(byteCh, 0);
		    range = intCh / 4294967296.0;
		    intCh = (uint)(range * 26);
		    return Convert.ToChar(intCh + 65);
	    }
	    public String getPassword()
	    {
		    return password;
	    }
	    public void setLength(int len)
	    {
		    length = len;
	    }
    }

Had a little hiccup converting the byte array to a character in the range I wanted, Im sure my solution is a bit sloppy. I wanted to use the secure random methods instead of just Random.Next(min, max).

[pickypg]

Lord Method Man, on 02 October 2012 - 19:21, said:

Had a little hiccup converting the byte array to a character in the range I wanted, Im sure my solution is a bit sloppy. I wanted to use the secure random methods instead of just Random.Next(min, max).

Note: you're still using Random.Next(...) in your main loop.

Also, you could probably simplify the code a lot with your bytes-to-number conversion being extracted into a separate function.

private char getCharacter(uint start, uint range)
{
    byte[] bytes = new byte[1];

    // realistically, given the expected ranges (not even a full byte), you could use a single byte
    rng.GetBytes(bytes);

    return Convert.ToChar(start + bytes[0] % range);
}

Any random lowercase character:

ch = getCharacter((uint)'a', 26);

Any random uppercase character:

ch = getCharacter((uint)'A', 26);

Any random number character:

ch = getCharacter((uint)'0', 10);